Working Paper 72

What Can We Learn from the Uganda Revenue Authority’s Approach to Taxing High Net Worth Individuals?

Jalia Kangave, Susan Nakato, Ronald Waiswa, Milly Nalukwago

and Patrick Lumala Zzimbe

January 2018

ICTD Working Paper 72

What Can We Learn from the Uganda Revenue Authority’s Approach to Taxing High Net Worth Individuals?

Jalia Kangave, Susan Nakato, Ronald Waiswa, Milly Nalukwago and Patrick Lumala Zzimbe

January 2018

2

What Can We Learn from the Uganda Revenue Authority’s Approach to Taxing High Net Worth Individuals? Jalia Kangave, Susan Nakato, Ronald Waiswa, Milly Nalukwago and Patrick Lumala Zzimbe ICTD Working Paper 72

First published by the Institute of Development Studies in January 2018 © Institute of Development Studies 2018 ISBN: 978-1-78118-413-4

This is an Open Access paper distributed under the terms of the Creative Commons Attribution Non Commercial 4.0 International license, which permits downloading and sharing provided the original authors and source are credited – but the work is not used for commercial purposes. http://creativecommons.org/licenses/by-nc/4.0/legalcode

Available from: The International Centre for Tax and Development at the Institute of Development Studies, Brighton BN1 9RE, UK

Tel: +44 (0) 1273 606261 Email: info@ictd.ac.uk Web: www.ictd/en/publications

IDS is a charitable company limited by guarantee and registered in England Charity Registration Number 306371

Charitable Company Number 877338

3

What Can We Learn from the Uganda Revenue Authority’s Approach to

Taxing High Net Worth Individuals?

Jalia Kangave, Susan Nakato, Ronald Waiswa, Milly Nalukwago and Patrick Lumala Zzimbe

Summary Wealthy people contribute a significant share of the total revenue collected through personal income tax (PIT) in high-income countries. This is not the case in most low-income countries, where the bulk of revenue from PIT is collected from people who are in formal employment, especially in the public sector. In most cases, PIT is collected by employers and remitted to the tax authority. The problem is not an absence of laws providing for the taxation of wealthy individuals (commonly referred to as High Net Worth Individuals (HNWIs)). Rather, these

laws are rarely implemented. On the one hand, this results in losses of income tax revenue, and, on the other, in severe inequity in the distribution of the tax burden. Successfully levying PIT on HNWIs requires a special organisational effort on the part of the tax authority. As far as we know, only three countries in Africa – Mauritius, South Africa and Uganda – have active systems in place to focus on the tax affairs of HNWIs. The Uganda Revenue Authority’s (URA’s) HNWI unit was set up in September 2015. Within the first year of its operation the unit increased revenue collection by UGX19 billion (USD5.5 million), and the proportion of wealthy individuals who filed income tax returns increased from 13 per cent to 78 per cent. These improvements were registered even before the URA audited any of the individuals.

A number of factors explain this success story. First, and most importantly, URA’s top management is actively engaged in and committed to the decision to tax these individuals. Second, once the URA obtained some information on potential HNWIs, it proceeded to act on its findings without waiting until it had in place a set of formal criteria for identifying these individuals. Most of the lessons are being learnt along the way. Third, the URA has not shied away from the fact that a large proportion of HNWIs are politicians or politically influential. To ensure that it deals with the political sensitivity of taxing these individuals, it has merged the HNWIs unit with its VIP unit, and placed both under the Public Sector Office, whose officials have experience in dealing with public figures. Fourth, there is close collaboration between the HNWI unit and the URA’s research department, which ensures that any recommendations made through research are tested in practice. Fifth, emphasis has been placed on officials in the HNWI unit having good communication skills. Sixth, people

identified as HNWIs are approached initially from a perspective of educating them and getting their commitment to pay some taxes, rather than undertaking harsh enforcement. Keywords: personal income tax; high net worth individuals; wealthy individuals; very important persons; politicians; public sector office; Uganda; criteria; Uganda Revenue Authority. Jalia Kangave is a Research Fellow at the Institute of Development Studies, and Capacity Building Manager of the International Centre for Tax and Development (ICTD). She also leads the Gender and Tax programme at ICTD. Susan Nakato is a supervisor in Corporate Performance, Monitoring and Evaluation at the Uganda Revenue Authority’s (URA) Research, Planning and Development Division. She is

4

one of the economists behind the African Tax Outlook publications, and a lecturer at Kyambogo University, Uganda. Ronald Waiswa is a supervisor in Research and Policy Analysis at URA’s Research, Planning and Development Division. He also heads research and training at Lida Africa.

Milly Nalukwago Isingoma is the Assistant Commissioner Research, Planning and Development at URA. She is also the URA head of Delegation to the East African Revenue Authorities Technical Committee, and a member of ICTD’s Centre Advisory Group. Patrick Lumala Zzimbe is an officer in Enterprise Risk Management at the URA’s Research, Planning and Development Division. He is a statistician and researcher in the areas of taxation, gender, food security, demographics and health, agriculture, and child labour in the mineral and energy sector.

5

Contents

Summary 3 Acknowledgements 6 Acronyms 6 Introduction 7

1 The URA’s approach to taxing HNWIs 9 2 Achievements of the URA’s HNWI/VIP unit 15 2.1 Profiles of HNWI/VIPs and an updated register 15 2.2 Filing of tax returns 15 2.3 Revenue collection 15 2.4 Improved voluntary compliance 16 2.5 Compliance audits of HNWIs 16 3 How is the URA’s approach to taxing HNWIs similar to or different from that of other tax authorities? 16

4 What lessons can the URA learn from other tax authorities? 18 4.1 Staffing requirements 18 4.1.1 Staff/taxpayer ratio 18 4.1.2 Staff development 19 4.1.3 Staff continuity 19 4.2 Encouraging voluntary compliance 20 5 What might other countries learn from the URA’s experience? 22 References 24

Tables Table 1 Tax register for PSO as at end of June 2017 10 Table 2 Analysis of potential HNWIs using the proposed criteria 14 Table 3 Revenue collection from HNWIs as at end June 2017 15 Table 4 Revenue contribution for HNWIs and VIPs for first half of FY 2015/16 15 Figures Figure 1 Structure of the PSO 11 Figure 2 Criteria for identifying HNWIs in Uganda 12 Figure 3 Compliance levels for HNWIs/VIPs 21

6

Acknowledgments We are grateful to the International Centre for Tax Development (ICTD) for the financial support that made this study possible. Special thanks go to Mick Moore and Wilson Prichard for their feedback on previous versions of the paper. We thank Adam Randon for his coordination and timely disbursement of funds. We also thank participants at the ICTD’s

2017 annual meeting in Entebbe, Uganda for their useful comments. We express our sincere appreciation to the Senior Management of the URA especially: Henry Saka, Commissioner Domestic Taxes; Silajji Kanyesigye, Commissioner Large Taxpayers Office; and Irene Bashabe, Assistant Commissioner Compliance Management. These leaders provided significant input in the study, including assisting with the development of the criteria for identifying high net worth individuals (HNWIs). We thank Tina Kaidu, Manager Research, Planning and Development Division for comments to previous drafts of the paper. Lastly, we thank the Public Sector Office/HNWI team, particularly James Etengu, Monica Tumukunde, Angus Arinaitwe and Elizabeth Aketch for the wealth of knowledge that they

shared with us.

Acronyms ATO Australian Taxation Office GDP Gross Domestic Product HMRC Her Majesty's Revenue and Customs HNWI High Net Worth Individual LTO Large Taxpayers Office OECD Organisation for Economic Cooperation and Development PAYE Pay As You Earn PIN Personal Identification Number

PIT Personal Income Tax POWG Privately Owned and Wealth Group PSO Public Sector Office SARS South African Revenue Service TIN Tax Identification Number UGX Uganda Shillings URA Uganda Revenue Authority USD United States Dollar VAT Value Added Tax VIP Very Important Person

7

Introduction The governments of many low-income countries now express a strong commitment to enhanced domestic resource mobilisation. They have, on average, greatly reduced their dependence on foreign aid over the last two decades. Domestic resource mobilisation was the first item on the Action Agenda at the 2015 international conference on Financing for

Development in Addis Ababa (United Nations 2015). At that conference a number of donor agencies and countries offered their support to strengthen the capacity of tax authorities to increase revenue collection (United Nations 2015; OECD 2015). Significant progress has been made. But there are still many challenges being faced, including the challenge of raising significant revenue through personal income taxes (PITs). The bulk of the revenue collected in many low-income countries comes from taxes on goods and services, notably Value Added Tax (VAT), and customs and excise duties. Income taxes often contribute a small percentage, and are largely paid by people in formal employment – in the public sector and a few large companies. PIT is estimated to contribute on average 10 per cent of GDP in high-income countries, compared to 2 per cent in sub-Saharan African countries (excluding South Africa) (Moore and Prichard 2017). Wealthy individuals in high-

income countries account for a significant share of PIT receipts, while they pay very little in low-income countries. In Uganda in FY 2014/15, taxes on goods and services (including VAT, excise and customs duties) constituted 66 per cent of total tax revenue, while PIT contributed 25 per cent (OECD et al. 2017).1 People in formal employment paid most of the PIT (68 per cent). Taxes such as rental income tax, which one would have expected to generate a lot of revenue given high and increasing property values, contributed only 1 per cent of total PIT receipts. Most of the remaining 31 per cent of PIT was collected from withholding taxes on imports, supplies to government, dividends, professional fees and non-residents. Ultimately, employees have borne and continue to bear most of the burden of PIT. This results in severe inequity in distribution of the tax burden. Various explanations have been given for the underperformance of income taxes in low-

income countries. These include weaknesses in tax administration, entrenched power structures, corruption and inappropriate transfers of tax systems from high-income to low-income countries (Keen 2012). Most, if not all, of these factors hinder PIT collection in Uganda. There is also arguably a historical explanation for the underperformance of PIT in the country. As is the case in the rest of East Africa, PIT in Uganda was originally designed for Europeans and Asians. When the tax was introduced in 1940, it was only imposed on non-Africans (Kwagala-Igaga 2013; Therkildsen 2006). Africans continued to be subject to the poll tax (imposed on adult men who owned huts), which was subsequently replaced by the graduated tax (a local government tax imposed on able-bodied men above the age of 18 and women in gainful employment). Even though the law was subsequently amended to impose PIT on all taxpayers irrespective of their race, in practice very few individuals were accustomed to paying it. Similarly, the Uganda Revenue Authority (URA) put little effort into

enforcing PIT, and relied mostly on withholding agents such as employers, and, to a lesser extent, companies (to withhold taxes from the payment of dividends).

Wealthy individuals in Uganda have historically paid little PIT. For FY 2013/14, for example, only 5 per cent of directors of the top taxpaying companies were paying income taxes, with some paying as little as USD5. Similarly, a sample of the top 60 lawyers in the country revealed that less than a third were remitting PIT for the period between FY 2011/12 and FY 2013/14. There was also a mismatch between payment of import duties and compliance with income tax obligations. Some individuals were paying billions of Uganda shillings in import duty, but not paying any income taxes. Lastly, an analysis of the compliance of 71 top

1 The figures relate to taxes collected by central government, excluding social security contributions.

8

government officials over the period 2011/12 to 2013/14 revealed that, although all of them had stakes in commercial enterprises, the majority were not paying PIT. Neither were the companies that they were associated with complying with their tax obligations (Kangave et al. 2016).

The under-taxation of wealthy individuals is not confined to Uganda or low-income countries more generally. Rhere are concerns that wealthy individuals in high-income countries do not pay their fair share in taxes because of issues such as aggressive tax planning and significant tax cuts (Piketty et al. 2014; OECD 2009). The tax affairs of wealthy individuals globally are increasingly being put under scrutiny for various reasons (OECD 2009). These individuals could contribute significantly to revenue collection. They are especially likely to engage in aggressive tax planning. The compliance behaviour of this group of taxpayers is likely to have a bearing on the behaviour of other taxpayers, and thus on the integrity of the tax system as a whole. Successfully taxing wealthy individuals requires a special organisational effort on the part of tax authorities. Tax authorities are increasingly targeting them, particularly in rich countries

(OECD 2009; ATAF 2017).2 The methods used vary from country to country. They include: setting up dedicated units to manage their tax affairs; paying special attention to the tax returns of larger companies in which HNWIs have significant shareholdings; and appointing officials in different departments within the tax authority to deal specifically with the affairs of HNWIs. To the best of our knowledge only three countries in Africa have specific active mechanisms for dealing with HNWIs: Mauritius, South Africa, and now Uganda. Uganda’s HNWI unit was set up in September 2015. At the end of the first year of operation, it had collected over UGX19 billion (USD5.5 million). Only 13 per cent of the 117 people initially identified as HNWIs had been filing income tax returns at the beginning of that year. By the end of the year, 78 per cent were filing.

What explains the URA’s success in such a short period of time? What should the URA do to ensure that this success is sustained? What can the URA learn from other tax authorities? And what can other tax authorities learn from the URA’s experience? To answer these questions, in Section 1 we document the steps that the URA has undertaken so far to tax HNWIs. We highlight the benefits that have resulted in Section 2. This is followed in Section 3 by an analysis of the similarities and differences between the URA’s approach and the approaches of three other tax authorities: Her Majesty’s Revenue and Customs (HMRC) in the UK, the South African Revenue Service (SARS) and the Australian Taxation Office (ATO). We selected the ATO because it has one of the longest histories of dealing with HNWIs. We chose SARS, because it is one of only three tax authorities in Africa that has an active system in place to focus on the affairs of HNWIs. Lastly, we selected HMRC because of the collaboration between HMRC and URA on a number of tax issues. We later

discovered that HMRC is the only one of the three that has a dedicated unit to deal with the affairs of HNWIs.3 While SARS and the ATO have systems in place to tax HNWIs, none of these are organised as special units. In Section 4, we discuss how the URA can improve the performance of its HNWI unit by learning from the experiences of HMRC, SARS and the ATO. We conclude our discussion in Section 5, by listing the lessons that other tax authorities can learn from the URA’s experience.

2 Taxpayer segmentation is a major feature of modern tax administration. Historically, tax authorities were structured

along tax types. However, tax type structures had many disadvantages, including promoting unhealthy competition among different tax offices to obtain taxes from the same taxpayers, failure to share information relating to the taxpayers, and high transaction costs for taxpayers who had to deal with different tax offices (Moore 2013). Segmenting

along taxpayer type has enabled revenue authorities to concentrate scarce resources on a small number of taxpayers who have complex transactions (e.g. large companies and HNWIs), and deploy less skilled staff on more routine tasks.

3 In addition to comparative analysis with the three revenue authorities, we conducted interviews with officials in the URA (within and outside Kampala), Ministry of Finance, Planning and Economic Development (Ministry of Finance), tax practitioners and a Member of Parliament. We also sat in on a meeting held by URA officials with prospective HNWIs.

9

1 The URA’s approach to taxing HNWIs Following a study conducted by the authors of this paper, in September 2015 the URA established an HNWI unit as part of the Large Taxpayer Office (LTO) in the Domestic Taxes Department. The unit was staffed with five officials (four tax officers and a supervisor). To begin their work, the officials generated a list of potential HNWIs, comprising directors of

large companies under the LTO and individuals whose wealth was publicly known. The list was scrutinised by top management in the Domestic Taxes Department, who made revisions on the basis of their own knowledge. Given that most of the individuals selected were not filing income tax returns, the URA did not at that point have records that it could use to ascertain the accuracy of the list. Further, the URA did not have any formal criteria for defining HNWIs. The initial list comprised 117 people. The HNWI unit staff then contacted them by telephone to set up appointments to talk about their tax affairs. The officials gathered as much information as they could from their databases about the economic assets and transactions of these individuals before the meetings. The URA was often represented by the supervisor of the unit and a tax officer at the meetings. In some cases the URA’s Commissioner General and Commissioner of Domestic Taxes accompanied the team. The meetings were intended to serve two broader purposes: to educate the taxpayers of their

rights and obligations, and to signal that the URA was looking into their tax affairs. The HNWI list included a number of politicians. Shortly before the 2016 presidential and parliamentary elections, the URA approached the Electoral Commission and requested that they should require a tax clearance certificate from the URA before accepting nominations from candidates for parliamentary seats. The Electoral Commission had no legal authority to insist on this requirement, and so it did not make the announcement itself. However, the URA, through the Domestic Taxes Department, made public announcements and sent out text messages and emails to members of the public stating in part: ‘If you intend to contest in the upcoming national elections, you need to obtain a Tax Clearance Certificate addressed to the Electoral Commission. To obtain a Tax Clearance Certificate, you will need to ensure that all your returns are filed and any tax due is paid’.4

Almost all candidates went to the URA to apply for the clearance. Before issuing clearance the URA insisted on evidence of filing returns, and required the candidates to make at least some payment towards what they owed. In the first month of introducing this rule, URA was able to raise approximately UGX380 million (USD110,000) from the aspiring members of parliament (Interview with URA official). In January 2017 the HNWI unit was moved from the Large Taxpayers Office to the Public Sector Office (PSO), where it was merged with the VIP unit that had been established in July 2015 to deal with the affairs of individuals who were considered by the URA to be politically influential.5 The HNWI/VIP unit currently has six staff, and responsibility for the tax affairs of 326 individuals. The decision to merge the units was driven by two main factors. First, a

number of individuals on the HNWI register were also VIPs. Of the original 117 HNWIs identified, 21 per cent were VIPs and 11 per cent former VIPs. Second, the URA observed that there were other similarities or overlaps: both groups tend to be politically or economically influential, and have very busy lifestyles. The skills required to interact with

4 <https://ugandaradionetwork.com/story/ec-explains -ura-tax-certificates-for-aspirants>. 5 Individuals dealt with by the VIP unit are divided into two broad categories. The first consists of government officials ,

such as the president of Uganda, the vice president, cabinet ministers, high ranking judicial officers, the speaker and deputy speaker of parliament, heads of political parties, heads of government institutions, the inspector general of police, the chief of defence forces, members of parliament that head committees , and executive directors of public

entities. The second category is made up of individuals who are not government officials but are considered to be influential, including kingdom heads, heads of professional associations, heads of business associations, religious leaders, and outspoken individuals such as influential journalists.

10

them and manage their affairs are quite similar (Interview with URA official). It was more efficient to handle them through one office. It is important to provide some background about the PSO itself to understand the rationale behind putting the newly merged unit under the PSO. It was established in September 2014

to handle the tax affairs of government ministries, departments, agencies and local authorities (Uganda Revenue Authority 2016). Previously, the affairs of most of these government organisations (except local government) were managed by the Large Taxpayers Office.6 The URA set up the PSO for various reasons:

Government employs a significant proportion of the population, so it was important to ensure that government was paying taxes on the income of its employees.

Government is also one of the biggest consumers of goods and services. However, many

of its suppliers were not registered with the URA.

The government regularly concludes contracts with local and foreign contractors in which it undertakes to pay all or part of the taxes due – or exempts transactions from taxes entirely. These contracts are not always known to the URA, particularly in cases where the contractors use different registration details from those contained in their Tax Identification Number (TIN) certificates.

The government is one of the biggest tenants and landlords in the country. A number of

those who let out their property to the government are not registered with the URA. Some of these property owners are government officials.

The PSO is thus tasked with a number of functions including:

Advising government organisations about their tax obligations, and educating them on the functions and departments of the URA.

Maintaining an up-to-date database of relevant government organisations.

Monitoring and ensuring that government organisations are up-to-date with their filing and payment obligations.

As at June 2017, the PSO was responsible for 948 taxpayers consisting of VIPs, HNWIs and government agencies (Table 1).

Table 1 Tax register for PSO as at end of June 2017

PSO register Number %

VIPs 209 22

HNWIs 117 12

Public sector organisations 622 65

Total 948 100

Source: URA database

The PSO is headed by a manager and has 25 staff members, including supervisors, tax officers and support staff.

6 The tax affairs of local government agencies were handled by local offices of the URA.

11

Figure 1 Structure of the PSO

There were a number of reasons for the decision to put the merged HNWI/VIP unit under the PSO:

Some of the taxpayers concerned are politically influential figures. It is important that they are dealt with by officials who have experience in dealing with politicians. There is also some history associated with this. Before the URA was established as a separate, semi-autonomous tax authority in 1991, there was a VIP unit under the Ministry of Finance (Interview with URA official). This unit was formed after a junior officer wrote a letter to the president, inviting him to go to the revenue office to make some declarations regarding his tax affairs. This raised a lot of concern. The Minister of Finance was required to explain the behaviour of the junior officers. A VIP unit was then established so that government officials above a certain rank would be dealt with by high-ranking tax

officers. However, the VIP unit was not transferred when the URA was established. It was not until 2015 that the URA decided it was important to have a special unit for these individuals to ensure that their tax affairs were handled with the level of confidentiality that the URA felt was necessary (Kangave et al. 2016).

The URA management felt that, for the Large Taxpayer Office (LTO) to operate more efficiently, it needed to be freed of responsibility for taxpayer segments that did not fit well within its operations. Since its establishment most of the work of the LTO had focused on large companies. Little attention had been paid to government organisations or HNWIs. The establishment of the PSO provided an opportunity to simplify the operations of the

LTO, by moving responsibility for government organisations and HNWIs to a more appropriate office.

Even before the HNWI/VIP unit was transferred to it, the PSO had registered considerable success in managing the affairs of a politically sensitive group of taxpayers (i.e. government organisations). The revenue collected by the PSO grew by 194 per cent in FY 2015/16, and 106 per cent in FY 2016/17. While the LTO collects more domestic revenue than any other office within the URA, the PSO is now second. It collected 4.5 per cent of domestic revenue in FY 2014/15, and 17.3 per cent in FY 2016/17.

Now that the HNWI/VIP unit is in place and various lessons have been learnt over the past

couple of years, the URA is working on formal criteria for identifying HNWIs. The authors of

Manager PSO

1 manager

Supervisor HNW/VIP, self-accounting

8 officers

Supervisor enforcement and reporting

8 officers

Supervisor compliance audits

4 officers

Admin/office attendant and driver

1 adm, 2 drivers & 1 office attendant

12

this paper worked closely with URA’s top management (particularly in the Domestic Taxes Department) and officials in the HNWI/VIP unit to develop the criteria. Since most potential HNWIs are not yet filing tax returns, and keeping in mind that many of them operate informally,7 we propose multiple income and wealth thresholds, comprising both core and non-core parameters. A core parameter is an indicator which is individually sufficient to

trigger classification as an HNWI. Any two non-core parameters generate the same result (Figure 2). Figure 2 Criteria for identifying HNWIs in Uganda

These parameters have been developed bearing in mind a number of factors: 1. Land and buildings. While property plays a central role in the economies of many low-

income countries (particularly in Africa), very little is collected from the taxation of property (Goodfellow 2015; Piracha and Moore 2015; Jibao and Prichard 2013). Land and buildings are the most common assets used by individuals to amass and retain wealth in Uganda (Murangira 2014). The revenue so far collected from taxing HNWIs in

Uganda demonstrates the central role of property in the country’s economy. Approximately 45 per cent of the taxes collected from HNWIs since the establishment of the unit came from property rental income. Similarly, out of all the revenue that has been

7 We use the term ‘informal’ cautiously to refer primarily to the challenges that countries such as Uganda face in

accessing taxpayer information or coordinating information that is in the hands of third parties. As we have noted elsewhere, there is ‘a large informal sector within the formal sector. By this we mean that there are many professionals

– both in the private and public sector – who may pay taxes on their employment income and/or importation of goods, but take advantage of the loopholes created by the informal sector to undertake additional lucrative commercial enterprises on which they do not pay taxes’ (Kangave et al. 2016: 12).

13

collected from VIPs so far, rental income tax constitutes 86 per cent (URA databases, 2017).

2. Shareholding in large companies or groups of companies with high turnover. Our research revealed that large companies whose tax affairs are managed by the LTO are typically controlled by a small group of individuals.8 Most of these individuals have

significant shares in more than one company. There are also individuals who control groups of companies that are under the Medium Taxpayers Office. It is often difficult to separate the private affairs of wealthy individuals from those of their companies (Interviews with URA officials and tax practitioners). Shareholders are often also the directors of their companies. Company transactions are often mingled with the private affairs of business owners. For example, company money is sometimes used to pay personal expenses (such as school fees for directors’ children and funding their private travel). These expenses are normally declared as general expenses in the business accounts. As one URA official stated: ‘The company is the individual. The signatory of the company bank account is one person. The board meetings are influenced by one individual. The same individual takes company money and uses it for personal effects such as constructing or renovating their homes’.

Many wealthy individuals invest through associates, such as children, spouses or pseudo business partners, who become the face of the company but in reality have little control over company decisions (Interviews with URA officials). It is thus important to scrutinise the affairs of the associates of HNWIs. Section 3 of the Income Tax Act defines an associate as any person who, not being an employee, acts in accordance with the directions, requests, suggestions or wishes of another person, whether or not they are in a business relationship. An associate could be a relative of the individual, a business partner, a partnership in which the person is a partner, a trustee of a trust under which the person benefits and a company in which the person – either alone or with associates – controls 50 per cent or more of the voting power in the company.

3. Loans and bank deposits. As one tax practitioner said: ‘You do not get a huge loan from a bank unless you have sufficient assets to show that you will be able to pay the money

back’. While wealthy individuals rarely declare their income to the URA and other government agencies, many of them provide banks with information relating to their assets in order to obtain loans (Interviews with URA officials and tax practitioners). Loans thus serve as a useful proxy for income and assets. Similarly, huge deposits on bank accounts are a useful indicator of wealth. In an interview in a URA office outside Kampala, staff told us of a farmer who approached the URA to apply for a tax identification number because he wanted to obtain a tax clearance certificate to bid for a government tender. One of the documents he submitted was a bank statement which indicated that he had UGX2.6 billion (USD742,857) in his bank account. The farmer had been in business for several years but had never registered for tax purposes. It is also common knowledge that several traders operating ‘small’ shops in the downtown Kampala area bank hundreds of millions of shillings on a daily basis.9 Yet the URA is

unable to establish the extent of their wealth because many of them do not keep proper books of accounts. The URA has made proposals to government requesting access to bank information. These proposals are still being discussed.

4. Publically known wealthy individuals. The identity of many HNWIs is publically known. This information can be gleaned from URA officials, members of the public, newspapers and lifestyle magazines. However, we categorise public knowledge as a non-core parameter because, on its own, this information is insufficient to confirm wealth. We also

8 We use the word ‘control’ in the manner in which it is used in Uganda’s Income Tax Act Cap 340, to mean having at

least 50 per cent of the voting power in the company. This control can be exercised either directly or indirectly through associates, other companies, partnerships or trusts.

9 The physical size of the shops is often small, with very few employees. However, for some of these traders turnover is in billions of Uganda shillings per year.

14

put professionals under this category, because some of the wealthy ones are publicly known.

5. Imports and exports. Just as land is an important form of amassing wealth in Uganda, international trade (particularly importation) is a lucrative business. As at FY 2016/17, exports constituted 11 per cent of GDP, while imports were 17 per cent of GDP (Ministry

of Finance 2017). Some individuals import goods worth billions of Uganda shillings and pay large amounts of money in import duties, but are not registered for income tax (Kangave et al. 2016). Information about exports and imports can be readily obtained from the URA’s Customs Department.

6. Farmers with high-value commercial forests, plantations and ranches. Uganda is largely an agricultural economy. Agriculture contributes 26 per cent to GDP (Uganda Bureau of Statistics 2015). While the majority of farmers operate on a small scale, there are a good number of individuals operating on a very large scale. An increasingly lucrative business for the wealthy is the maintenance of commercial forests, which are sold for carbon credits. By 2013, for example, privately owned forests occupied 2.3 million hectares (NFA 2013). Most of this forest land is owned by a few individuals. Similarly, in regions such as western Uganda, livestock and banana farming are a major source of income and wealth.

A URA official in an office outside Kampala informed us, for example, of an individual who owns 6 ranches, each with over 6,000 cattle. Another interviewee from the Kampala Capital City Authority, referring to the same individual said, ‘Each ranch of this individual is worth over a billion shillings’.

Having developed these criteria, URA management had one more question for the researchers: ‘Based on these criteria, who are the HNWIs?’ The researchers checked various URA databases using the criteria to answer this question. We found many potential HNWIs who are not among the 117 originally identified. Many HNWIs are VIPs (Table 2). Table 2 Analysis of potential HNWIs using the proposed criteria

Details Yardstick Potential

HNWIs

Already

on HNWI register

Already on

VIP register

Not on

HNW/VIP register

Shares Shareholders in companies with turnover greater than UGX50 bn or USD14.29 mn

493 13 5 475

Multiple shareholding in companies with

turnover ranging from UGX15 to 50 bn or USD4.29 to 14.29 mn

36 4 0 32

Commercial

buildings

Generates rental income greater than

UGX500 mn or USD0.14 mn per year

163 49 010 114

Land Land transactions worth UGX1 bn or

USD0.29 mn and above in the last 5 years

197 8 2 187

Motor vehicles Vehicle(s) with total value above UGX500 mn or USD0.14 mn both individually or in

aggregate

109 12 7 90

Imports and

exports by individuals' TINs11

Imports worth UGX500 mn or USD0.14 mn

per year

83 3 0 80

Exports worth UGX500 mn or USD0.14 mn

per year

22 0 0 22

Source: Authors’ computations based on URA databases

10 Most of the revenue from VIPs (86%) is from rental income tax, but none of them declared rental income greater than

UGX500 mn. 11 Some individuals import/export using company TINs. This information relates to those importing/exporting using

personal TINs.

15

2 Achievements of the URA’s HNWI/VIP

unit While the HNWI unit has only been established for two years, it has already registered impressive success. Below we list its achievements. 2.1 Profiles of HNWI/VIPs and an updated register

The URA paid very little attention to the tax affairs of individuals before the unit was established. The LTO had a register of 737 taxpayers, out of whom only 17 were individuals (Kangave et al. 2016). The majority of those individuals declared only the income they obtained from employment. There is now a register of 117 HNWIs and 239 VIPs. With the newly compiled register, the HNWI team has revised the registration details of all the individuals to include other sources of income. 2.2 Filing of tax returns There is great improvement in the filing of income tax returns by HNWIs. Many of the individuals had never filed an income tax return. Only about 13 per cent of the individuals in the LTO were filing returns before the HNWI unit was set up. At least 78 per cent of HNWIs and 65 per cent of VIPs were filing returns by FY 2015/16.

2.3 Revenue collection The HNWI unit was established in September 2015. By the end of June 2016, the unit had collected over UGX19 billion (USD5.5 million) in rental tax, personal income tax, VAT and stamp duty. This was a significant increase when compared to the UGX1.35 billion (USD390,000) that was collected from individuals in the LTO in FY 2014/15. In total, UGX40.05 billion (USD11.44 million) had been collected as at June 2017. Table 3 Revenue collection from HNWIs as at end June 2017

Tax head FY2015/16 FY2016/17 Grand total

Rental 9,678,619,886 7,306,044,119 16,984,664,005

Income tax 5,833,116,720 5,793,580,255 11,626,696,975

VAT 3,297,085,469 7,291,237,447 10,588,322,916

PAYE 370,331,167 484,931,799 855,262,966

Total 19,179,153,242 20,875,793,620 40,054,946,862

Source: URA database

In the first half of 2015/16, when the unit had just started its operations, 45 per cent of the taxes collected from HNWIs were from rental income, while for VIPs rental income taxes contributed

86 per cent of total collection. Table 4 Revenue contribution for HNWIs and VIPs for first half of FY 2015/16

HNWIs VIPs

Rental Income tax 45% 86%

Individual income tax 33% 13%

VAT 20% 1%

PAYE 2% 0%

Total 100% 100%

Source: Authors’ computations based on 2015/16 half year revenue collection statistics

16

2.4 Improved voluntary compliance In addition to increased revenue and improvement in filing of returns, the URA has noticed an improvement in the attitude of HNWIs/VIPs towards paying taxes. Specifically, they have noticed that, for some, non-compliance was due to a lack of understanding of their tax obligations. Similarly, some of the individuals approached appear to have influenced the

compliance of other individuals within their networks. For example, one of the business people that the URA met in the initial stages of the operation of the unit informed URA of their wider business network, some of whom have now improved their compliance. Similarly, one of the prominent pastors in Kampala city who is on the VIP list was reported as having told his church congregation that they should start paying their taxes. This was after the HNWI/VIP team paid him a visit and sensitised him about his taxpaying obligations. As HNWI/VIP unit officials pay more visits to taxpayers and communicate not just the rights and obligations of taxpayers, but also a willingness on the part of the URA to assist taxpayers in managing their tax affairs, some taxpayers get a more positive perception of the URA. 2.5 Compliance audits of HNWIs In FY 2016/17, the URA conducted comprehensive audits on the accounts of five individuals

on the HNWI register. Over UGX1.48 billion (USD490,000) was assessed and agreed upon with the taxpayers, as shown below.

Tax head Audit yield (UGX)

Income tax 760,999,221

VAT 510,388,567

Rental 213,370,720

Total assessed and agreed 1,484,758,508

3 How is the URA’s approach to taxing

HNWIs similar to or different from that of

other tax authorities? The URA’s approach to taxing HNWIs is similar to that of the British, South African and Australian tax authorities in two main respects: 1. All three tax authorities had at some point in time special units dedicated to handling the

affairs of HNWIs. The HMRC’s HNWI unit was established in 2009 and is still in operation (National Audit Office 2016). The South African Revenue Service (SARS) operated an HNWI unit between 2009 and 2014, after which the unit was closed because SARS restructured the whole organisation to operate along functional lines (Interview with

SARS officials, 2016).12 The Australian Taxation Office (ATO) established a High Wealth Individuals (HWI) Taskforce in 1996 to monitor and conduct compliance reviews of HNWIs (ATO response to questionnaire, 2016). The taskforce was disbanded in 2011. The affairs of HNWIs are now handled through ATO’s Privately Owned and Wealth Groups (POWGs), which deal generally with wealthy individuals and the companies they

12 Organisation by function is one of the modern systems of tax administration. Under this structure the revenue authority

is organised around different functions, such as taxpayer registration, audit and investigation, legal affairs, investigation and debt collection (see Moore 2013).

17

control. While SARS and ATO no longer have special units that strictly handle HNWI affairs, they still have systems in place to monitor the affairs of these individuals closely.

2. Many indicators of wealth (e.g. property, shareholding, professional services and luxury items like cars) used by the URA are similar to those used by the other three tax authorities. One indicator in particular is worth mentioning. The URA’s proposed criteria

of including company shareholding as a core parameter for identifying HNWIs is similar to Australia’s approach of handling the affairs of wealthy individuals through POWGs. The ATO resorted to this strategy because it observed that wealthy individuals controlled large complex groups of entities, and it was difficult to divorce the individuals from the entities. The ATO thus takes a ‘group wealth approach’, which enables it to assess the wealth of HNWIs through some of the most successful businesses in the country. For example, 379 HNWIs have controlling interests in more than 50 entities per individual (ATO response to questionnaire, 2016). The Ugandan situation is likely similar.

There are also a number of differences between the URA’s approach and that of the other tax authorities:

1. Unlike the URA, the other three tax authorities use a single threshold based on aggregate income and/or wealth. In the URA’s case there are multiple thresholds, which depend on the parameter that is being used. The URA uses multiple thresholds for three main reasons. First, information about economic activities is sometimes not officially recorded. It is quite common to have wealthy individuals whose businesses are not registered with government organisations. Second, even where such information is recorded, the information systems in the different government organisations rarely interface. A few government organisations, particularly outside Kampala, still rely on manual records. This makes it difficult to collate the information. Third, many transactions are undertaken using cash, and are thus difficult to trace. In the end, the URA is operating in an environment in which access to information poses a great challenge. This makes it difficult to aggregate the wealth/income of a taxpayer, at least at present.

2. As mentioned above, instead of having dedicated units for handling the affairs of HNWIs, SARS and ATO now manage the affairs of these individuals using different mechanisms. In SARS, following a restructuring of the organisation in 2014, the affairs of HNWIs started to be handled by the various functional departments (e.g. taxpayer services, accounts, compliance, case selection and audit). Instead of having a separate unit for HNWIs, there are tax officials in each of the new departments who are assigned to deal specifically with HNWIs. There is also a function known as the Taxpayer Strategy, which is tasked with drafting policies for enhancing the compliance of HNWIs, and ensuring that there is coordination of the affairs of these individuals between the different departments (Interview with SARS officials, 2016). In the ATO, the affairs of wealthy individuals are dealt with across compliance teams that handle the broader POWG population.13 While there is no dedicated unit to deal solely with the individuals, the POWG monitors their

affairs closely and reviews all high-risk wealthy individuals over a three-year cycle.

3. Unlike the other three tax authorities, URA’s approach explicitly includes politicians as HNWIs. There are no politicians on the HNWI register in South Africa (Interview with SARS officials, 2016). A separate unit (the Politically Influential Unit) deals with politicians. HNWIs for SARS’s purposes are thus private individuals. We asked the ATO: ‘Have you had to deal with HNWIs who are influential public figures, such as politicians? If so, what has been your experience dealing with them? How have you handled any challenges resulting from their political influence?’ The ATO responded: ‘Yes, high profile taxpayers are treated the same way as ordinary taxpayers – impartially and according to

13 POWGs consist of three taxpayer categories: private groups (economic groups with a turnover of more than AUD2

million); wealthy Australians (resident Australian individuals who, together with their associates, control wealth of between AUD5 million and AUD30 million); and high wealth individuals (resident Australian individuals who, together with their associates, control wealth of at least AUD30 million).

18

the law’. It is one thing to have a provision in the law that provides for the taxation of all individuals and quite another to enforce that provision. Success in implementing laws differs from one country to another. Enforcement is even more precarious when powerful individuals, such as politicians, are being targeted.14 URA found that unless it deliberately

included politicians in its strategy, it would be difficult to collect taxes from them.

4 What lessons can the URA learn from other

tax authorities? There are various lessons that the URA can learn from HMRC, ATO and SARS. We divide these into two broad categories: staffing requirements and encouraging voluntary compliance. 4.1 Staffing requirements

There are three issues relating to human resources that the URA needs to address: the ratio of tax staff to taxpayers, staff development and staff continuity. 4.1.1 Staff/taxpayer ratio

The HNWI/VIP unit has 6 staff members serving a total of 326 individuals. This puts the staff to taxpayer ratio at 1:54. With the newly developed criteria, the number of HNWIs is certain

to increase. The current ratio is already relatively high when compared to HMRC’s HNWI unit, for example. As at 2015/16, HMRC’s unit had approximately 380 staff members handling the tax affairs of 6,500 HNWIs, bringing the ratio of staff to taxpayers to 1:17 (National Audit Office 2016). In addition, each HNWI in HMRC has a customer relationship manager who is responsible for understanding the behaviour of the individual, the risks associated with them and their business transactions. As at 2015/16, there were about 40 customer relationship managers. We are aware that the shortage in human resources is not unique to the HNWI/VIP unit of URA. The organisation is generally understaffed. In a recent study of 15 tax authorities conducted by the African Tax Administration Forum, URA was ranked as the second most understaffed tax authority (after the Burundi Revenue Authority), with the population to tax

administrator ratio being over 6,000:1 (African Tax Administration Forum 2016).15 Almost all the other tax authorities had a ratio of less than 4,000:1. Internal URA records reveal that in FY 2015/16 the overall staff to taxpayer ratio was 1:373. It was worse for the domestic tax auditors, where the number of auditors to taxpayers stood at 1:4,319. While staffing is clearly a challenge in the URA generally, there is a strong case for increasing its human resources given the potential for revenue increase demonstrated by the HNWI/VIP unit. Initially, this can be achieved through staff movements from other departments. However, there will be a need to recruit externally in the long term given the overall staff shortage.

14 e.g. In November 2016 Uganda’s Members of Parliament passed a bill in which they exempted themselves from paying

income taxes on their allowances, despite the fact that a court had just ruled that those allowances should be subjected to tax. See Parliament of Uganda, ‘Parliament passes Income Tax Bill, justifies tax exemptions on allowances’ <http://www.parliament.go.ug/index.php/about-parliament/parliamentary-news/1037-parliament-passes-income-tax-bill -

justifies-tax-exemptions-on-allowances>. 15 This ratio takes into account both potential and registered taxpayers.

19

4.1.2 Staff development While the manager of the PSO and the supervisor of the HNWI/VIP unit both have vast experience in tax issues and working in the URA,16 the majority of officials in the unit are relatively junior. Two of the officers have seven years of experience working in the URA,

while the remaining four are previous graduate trainees who have less than five years of experience (Interview with URA officials). The new focus on HNWIs is likely to result in these individuals getting tax advisors to help them manage their affairs. The URA will need to match this expertise. Some of this expertise can be obtained within the URA.17 One of the reasons why SARS has achieved considerable success in taxing HNWIs is because it uses technically competent auditors who are attached to the cases of specific HNWIs (Interview with SARS officials, 2016). Some of these auditors were recruited from the private sector. The URA should look into recruiting experienced tax advisors from audit firms and law firms. There is also a need for training through a mixture of on-the-job training, short courses

geared toward addressing specific needs, informal networking events and secondments to other revenue authorities (OECD 2009). In SARS, for example, junior auditors were exposed to complex tax planning schemes as soon as they started working with the HNWI unit to build the capacity of lower-level staff (Interview with SARS officials 2016). In addition, these officers frequently met their more experienced counterparts to discuss the various cases being handled by the unit. In addition to on-the-job training, URA needs a well-structured programme for cadre development, including using ex-URA officials to pass on their skills to more junior staff (Interview with officials in the Ministry of Finance). Lastly, the OECD has recommended that where a revenue authority is not able to hire from the private sector, regular dialogue and interaction with the advisors of HNWIs can increase officials’ understanding of the business of the taxpayer (OECD 2009).

4.1.3 Staff continuity One of the challenges to building capacity in the URA is the constant transfer of officials from one office to another (Interviews with URA officials). Staff members are transferred just as they are beginning to understand their new roles. In some cases resources have been spent on external training of an official for a specific role, and the training is not directly applicable to their new role. Since the establishment of the HNWI unit, the supervisors have been changed three times, and some officers have been moved to other offices. This is a high turnover considering the unit is only two years old. However, one senior official explained that rotation within the URA occurs for various reasons. Sometimes, there is a need for more human resources in a particular department, but the URA does not have the funds to recruit new staff. At other times, officials are moved because they have served at a station for so

long that they are at risk of being compromised. It should also be remembered that there was a lot of experimenting with different ideas in the initial stages of establishing the HNWI unit; this is reflected in the staff changes and even the merger of the VIP/HNWI units. However, for the purposes of building capacity, it will be important in future that there is some continuity in the staff attached to the unit. Movements do not only affect capacity building. They also disrupt the relationship that has been built between URA staff and HNWIs.

16 The manager of the PSO has been in the URA since it was established, and has served in various departments and

capacities. Before URA’s establishment, he worked in the revenue office in the Ministry of Finance. The supervisor has

worked with the URA for 11 years, has experience in dealing with public officials, and is firm in her approach. 17 While the majority of the URA’s staff members (41%) have less than five years of working experience, there are also a

significant number that have long-term working experience: 20% have worked for between 5-9 years; 28% between 10-19 years and 11% have at least 20 years’ experience. (These statistics relate to working with the URA. The URA has been the only employer for many officials).

20

The risk of being compromised as a result of constant interaction between tax officials and HNWIs should be monitored. In 2017, the UK’s Public Accounts Committee raised questions about HMRC’s seemingly cosy relationship with HNWIs, arguing that having customer relationship managers gave the impression that HNWIs receive favourable treatment (Committee of Public Accounts 2017). HMRC responded that while HNWIs appeared to be

treated favourably, they also received more scrutiny than most other taxpayers. To manage the risks associated with taxpayers or their agents working closely with customer relationship managers, the HMRC puts a number of checks in place:

HNWIs are rotated between teams

Customer relationship managers work with a team of 17 other staff members

Customer relationship managers are not solely responsible for signing off key decisions such as whether to open an inquiry into an HNWI. These decisions are signed off by a staff member outside the team

Disputes relating to an HNWI’s case are handled by the HMRC’s dispute settlement mechanisms; customer relationship managers who have been central to the enquiry cannot approve the settlement.

The URA can learn from HMRC’s approach by putting mechanisms in place to check the independence of officials working in the unit, while ensuring that there is continuity of staff. 4.2 Encouraging voluntary compliance To sustain the success of the HNWI/VIP unit, the URA needs to put mechanisms in place to encourage voluntary compliance. Many tax authorities in high-income countries aspire to achieving what is known as ‘cooperative compliance’ in their dealings with HNWIs. Cooperative compliance encourages regular interaction between the revenue authority and HNWIs (or their advisors), so that there is a consistent flow of information between the taxpayers and the tax authority, and mutual awareness of the issues relating to the taxpayer (OECD 2009). The goal of cooperative compliance is to develop a relationship of trust between the two parties, by ensuring that sensitive personal information relating to the

taxpayer is safeguarded, that there is a well-developed legal framework in place (and taxpayers are informed of changes to this framework), and that the tax authority is impartial, responsive and competent in the manner in which it deals with the affairs of the taxpayer. Compliance, particularly among this group of taxpayers, can be significantly improved by providing them with timely and comprehensive guidance. To encourage a good working relationship between HMRC and HNWIs, and to increase voluntary compliance, HMRC ensures that customer relationship managers try to resolve issues before HNWIs submit their tax returns (National Audit Office 2016). These managers give HNWIs the HMRC’s position on a specific issue before the HNWI engages in a particular activity. If HMRC and the taxpayer fail to agree on a position, the HMRC undertakes a formal inquiry. If it finds that the HNWI was careless in their submissions, it

may penalise the individual. However, sometimes penalties are suspended in order to encourage taxpayers to avoid similar mistakes in the future. In cases of suspected fraud, the matter is passed on to a specialist team, which determines whether there will be a criminal or civil investigation. How ATO treats HWIs depends on their behaviour. As is the case with HMRC, the ATO engages with these individuals before they file returns, in order to give them an opportunity to discuss their affairs and get ATO’s opinion (ATO response to questionnaire, 2016). Where individuals are perceived to be low risk, ATO’s approach is to educate them, guide them and provide them with the support that they need in filing their returns. If the ATO perceives an individual as having some degree of risk of non-compliance, it takes steps to manage the

21

risk. For example, it may decide to secure face-to-face meetings in which an individual is informed of what the ATO knows about their tax affairs, and given an opportunity to respond. Following this initial meeting, the ATO gives the individual a contact person who will deal with any subsequent issues. The inquiries may result in an individual being rated as high risk, in which case the ATO explores a variety of options for dealing with them, including audits,

investigations by the crime unit, and working with law enforcement in the case of serious crime. Lastly, SARS takes a similar approach through its motto of education, service and enforcement (Interview with SARS officials, 2016). SARS aims to engage with all taxpayers by providing them with the required education and services to support their compliance. When education and service provision fail, enforcement action follows. So far, the URA is moving in the right direction. Officials in the HNWI/VIP unit informed us in an interview that:

The VIP unit was formed as a sort of public relations arm. Our priority is to change the

attitude of the people by giving them free advice. We do not begin by presenting ourselves as people who have come to enforce, even though that is our ultimate objective. When we go to VIPs, we tell them how the unit was formed and why it was formed. We inform them about their rights and obligations, and then tell them the information that we have about their commercial activities.

This approach explains, at least in part, the success that the URA has achieved so far. However, as the graph below shows, there are risks of reduced compliance, particularly among VIPs. Figure 3 Compliance levels for HNWIs/VIPs

Source: URA databases

Cooperative compliance does not mean that a tax authority should relax its rules on non-compliance. There should be mechanisms in place to penalise uncooperative taxpayers. The

10.22

8.88

9.84

0.31 0.21 0.18 -

2.00

4.00

6.00

8.00

10.00

12.00

FY1516_First Half FY1516 second half FY1617 first half

Half Year Domestic Revenue collections UGX bns

HNWI VIP

22

integrity of the HNWI/VIP unit will thus be partly tested by its ability to enforce against those taxpayers who fail to comply with their obligations, even after the URA has educated them. One other issue that needs to be addressed is how to deal with past non-compliance. There are generally two kinds of taxpayers: those who will always be non-compliant, and those who

are willing to be compliant but are uncertain of the consequences of disclosing non-compliance (OECD 2009). For the first category, very little – if anything – can be done by the tax authority to encourage voluntary compliance. Enforcement through penalties is often the only option for this category. The second category can improve its compliance if there is some kind of assurance that the disclosure of non-compliance will not result in criminal charges, reputational damages (when non-compliance is publicised), future intensive audits, and inability to settle the interest and penalties associated with past non-compliance. One way of dealing with the second category is by putting in place a voluntary disclosure mechanism which consists of: a) Publishing a guide on the procedure to be followed when voluntarily disclosing past non-

compliance

b) Ensuring that the information disclosed by taxpayers is kept confidential by putting in place a range of mechanisms including special legislative secrecy provisions, designating specific officials to have access to the disclosed information and operating a disclosure system via a banking system

c) Publishing broad principles on the procedures that will be followed once information is disclosed, to assure taxpayers that disclosure of non-compliance does not necessarily result in the risk of more intensive monitoring and auditing of the business of the taxpayer

d) Providing clarity on the penalties, interest or criminal prosecution, if any, that may result from the declaration and stating mitigating factors

e) Providing a dedicated phone number that those wishing to disclose can call to have initial discussions with the revenue authority without the requirement of disclosing the taxpayer’s identity.

(OECD 2009: 65-67) Voluntary disclosure programmes in South Africa, are clearly stipulated in legislation and publicly announced (Interview with SARS officials, 2016). SARS believes that these programmes have merit, and bring within the tax net cases of non-compliance that SARS may never have discovered, given the limited resources available for investigating non-compliance. The URA should look into the merits of having voluntary disclosure mechanisms. However, if these mechanisms are to be introduced the mechanisms should be available to all categories of taxpayers, and not just HNWIs, to prevent a backlash from other taxpayers.

5 What might other countries learn from

URA’s experience? The URA’s experience provides many lessons, particularly for tax authorities in Africa. We summarise these below:

To tax a segment of the population that is economically and politically influential, it is important to have support from top management of the revenue authority. Both the URA’s Commissioner General and the Commissioner for Domestic Taxes have been committed

23

to ensuring that the decision to tax HNWIs is implemented. In some cases, these leaders have personally attended initial meetings with people identified as HNWIs. Other senior staff, at the level of Commissioner and Assistant Commissioner, have been involved in developing the criteria for identifying HNWIs.

The identity of most HNWIs is publicly known. It has not been difficult for the URA to

identify many of the most prominent. Initial information is cross-checked to ensure that people are not wrongly labelled as HNWIs.

Sometimes it is important to simply start with the little information available, without waiting until one has good definitions and comprehensive lists. The learning process is continuous, and definitions and lists can be revised along the way.

With will and intelligence, a tax authority can go ahead and make real progress without

focussing too much on what other countries do.

Technical expertise is important. But communication skills are even more important in a country where the majority of HNWIs engage more in tax evasion than avoidance, and where they wield a lot of political influence. Officials in the HNWI unit are chosen for their ability to balance between being assertive and being respectful. They are also able to communicate tax matters in a simplified manner.

Initially, emphasis is placed on educating taxpayers about their rights and obligations.

Enforcement comes later.

Identifying HNWIs requires close collaboration between the research function and the operational function. This allows for research findings to be tested in real time, and for the research, in turn, to be informed by what is happening in practice.

Government is one of the biggest consumers of good and services. HNWIs are likely to

be engaged – directly or through their associates – in the provision of goods and services to government. The URA’s Public Sector Office, which monitors the award of government contracts, has been a good source of information on contracts entered into between the government and these individuals.

Elections may have an effect on revenue collection. HNWIs who are politicians are more likely to pay their taxes during election periods. Whether these payments continue after election periods cannot be guaranteed.

24

References African Tax Administration Forum (2017) African Tax Outlook 2017, second edition, ATAF

—— (2016) African Tax Outlook 2016, first edition, ATAF

Committee of Public Accounts (2017) Collecting tax from high net worth individuals, thirty-

sixth report of Session 2016-2017, House of Commons

Goodfellow, T. (2015) Taxing the Urban Boom: Property Taxation and Land Leasing in Kigali

and Addis Ababa, ICTD Working Paper 38, Brighton: International Centre for Tax and

Development Jibao, S. and Prichard, W. (2013) Rebuilding Local Government Finance After Conflict: The

Political Economy of Property Tax Reform, in Sierra Leone, ICTD Working Paper 12,

Brighton: International Centre for Tax and Development Kangave, J., Nakato, S., Waiswa, R. and Lumala Zzimbe, P. (2016) Boosting Revenue

Collection through Taxing High Net Worth Individuals: the Case of Uganda, ICTD

Working Paper 45, Brighton: International Centre for Tax and Development

Keen, M. (2012) Taxation and Development – Again, IMF Working Paper WP/12/220,

International Monetary Fund Kwagala-Igaga, D. (2013) When More is Less: An Analysis of Reforms in the System of

Direct Taxation of Profits from Business Activity in Uganda, dissertation for the Degree

of Doctor of Philosophy, School of Law, University of Warwick, mimeo Ministry of Finance, Planning and Economic Development (2017) Background to the Budget:

Financial Year 2017/2018

Moore, M. (2013) Obstacles to Increasing Tax Revenues in Low Income Countries, ICTD

Working Paper 15, Brighton: International Centre for Tax and Development —— and Prichard, W. (2017) How can Governments of Low-Income Countries Collect More

Tax Revenue?, Working Paper 70, Brighton: International Centre for Tax and

Development Murangira, J. (2014) ‘A critical analysis of the right to equality before the law in property rates

tax law and practice in Kampala capital city authority’, Merit Research Journal of Art, Social Science and Humanities 2(8): 115-120

National Audit Office (2016) HMRC’s approach to collecting tax from High Net Worth

Individuals HC 790 SESSION 2016-17, HM Revenue and Customs

National Forest Authority (2013) The National Forest Plan 2011/12 – 2021/22, Ministry of

Water and Environment Directorate of Environmental Affairs, Uganda OECD (2015) Strengthening Tax Systems to Mobilise Domestic Resources in the Post-2015

Development Agenda Element 11, Paper 2, OECD

—— (2009) Engaging with High Net Worth Individuals on Tax Compliance,

<http://www.keepeek.com/Digital-Asset-Management/oecd/taxation/engaging-with-

25

high-net-worth-individuals-on-tax-compliance_9789264068872-en#page1>, accessed 4 May 2015

OECD/ATAF/AUC (2017) Revenue Statistics in Africa 2017, OECD Publishing, Paris

<http://dx.doi.org/10.1787/9789264280854-en-fr>

Piketty, T., Saez, E. and Stantcheva, S. (2014) ‘Optimal Taxation of Top Labor Incomes: a

Tale of Three Elasticities’, American Economic Journal: Economic Policy 6.1: 230-271

Piracha, M. and Moore, M. (2015) Understanding Low Level State Capacity: Property Tax

Collection in Pakistan, ICTD Working Paper 33, Brighton: International Centre for Tax

and Development Therkildsen, O. (2006) The Rise and Fall of Mass Taxation in Uganda: 1900 – 2005, DIIS

Working Paper 25, Copenhagen: Danish Institute for International Studies Uganda Revenue Authority (2016) ‘Domestic Taxes Department: Proposed Structural

Adjustments - July 2016’ (unpublished). Uganda Bureau of Statistics (2015) 2015 Statistical Abstract, Kampala, Uganda: Uganda

Bureau of Statistics United Nations (2015) Resolution Adopted by the General Assembly on 27 July 2015: Addis

Action Agenda of the Third International Conference on Financing for Development

Agenda Item 18

International Centre for Tax and Developmentat the Institute of Development StudiesBrighton BN1 9RE, UK T: +44 (0)1273 606261F: +44 (0)1273 621202E: info@ictd.acwww.ictd.ac